NJ Server Farms Remake the US Financial Markets

1sockchuck writes "The engine of Wall Street has shifted from the stock exchange floor to data centers in New Jersey, where computer-driven trading now accounts for 56 percent of all trading activity, according to the New York Times. 'While this Tron landscape is dominated by the titans of Wall Street, it affects nearly everyone who owns shares of stock or mutual funds, or who has a stake in a pension fund or works for a public company,' the Times writes. 'For better or for worse, part of your wealth, your livelihood, is throbbing through these wires.' There are also photos of the data centers powering the high-speed trading operations, while 60 Minutes has video of a huge new 'liquidity center' run by the NYSE."

Comments

Throbbing money

[Anonymous Coward]

'For better or for worse, part of your wealth, your livelihood, is throbbing through these wires.'

Can pennies throb?

Re:Throbbing money

[migla]

>Can pennies throb?

Sure. It's a widely used idiom, popular in phrases such as "The sweaty lumberjack lustfully thrust his throbbing pennies deep into the moist slot of the pink piggy-bank."

Re:Throbbing money

[Anonymous Coward]

Apparently those wires are actually penises. They are probably about as thin as CmdrTaco's micropeen as well.

Re:Throbbing money

[iluvcapra]

That cromulent metaphor embiggens this thread.

Re:Throbbing money

[the_hellspawn]

So, that's how you save. Intriguing!

short term skimming

[FuckingNickName]

And nothing of value was gained.

Re:short term skimming

[purpledinoz]

Exactly. Goldman Sachs and JP Morgan earn a huge chunk of their profit from high-frequency trading. This profit must come at an expense of someone else (like regular stock holders). In my mind, this is legal theft.

Re:short term skimming

[certron]

This is why the markets creep up on almost no volume and everyone wonders why stocks drop so quickly when someone sells a large enough block of stocks. These programs are the reason for the 'flash crash' and not the ridiculous 'typo by a trader' explanation. The argument of 'providing liquidity' doesn't really seem like it has that much value for a normal investor. It's robots all the way down!

Re:short term skimming

[hitmark]

and each year it will get worse, as some 3 billion and change (with compound interest) will need to find a place to be invested to keep the growth rate of the global economy that one have come to expect since the industrial revolution.

Re:short term skimming

[Yvanhoe]

Indeed. Actually, making the market trade with a period of one day, and randomizing the priority of orders arriving would sanitize a lot of things.
I have yet to see a good argument against that. I mean one that doesn't say things like "pschhh, you know nothing about economy, let us manage this thing". Sorry guys. We paid for your dumb errors in the subprime crisis, so now that you are creating speculation on yet another imaginary value (physical closeness to the servers of the stock exchange ? Really ? Changes in the value of a company on the millisecond scale ? Are you serious ? ) you better show your whole scheme.

Re:short term skimming

[Sponge Bath]

I would guess most of the profit comes from people on the losing side of a high-frequency trade. Buy and hold investors should be largely immune from short term churn.

Re:short term skimming

[0100010001010011]

With NTP, why don't they offer 2 buys/sells a minute? That lets everyone look at something, decide if they want to buy or sell, and it stops all this instatrading from happening.

On top of that, it seems that this lightning fast trading works great and they're happy if they're making money. If something cascades into failure (like it did earlier last year, or was it '09?) then they just say 'oops, do over'. Meanwhile some people were caught up that were cashing out their pensions or transferring funds between accounts.

I mean, they wouldn't make as much, but it'd be fair to the common person.
-
OR, the other suggestion that I saw would be to tax trades inversely proportional to how long they're held.
1 minute: 90%
1 hour: 80% .. .. .. ..
20 years: 5%
40 years: 1% (people that actually it as investment).

Re:short term skimming

[raw-sewage]

Exactly. Goldman Sachs and JP Morgan earn a huge chunk of their profit from high-frequency trading. This profit must come at an expense of someone else (like regular stock holders). In my mind, this is legal theft.

I see this mantra repeated often around here, but I'm not so sure it's entirely true. First, what is a "regular stock holder"? On one end, there are small-time, buy-and-hold investors such as myself; on the other end, there are big institutional investors who manage massive portfolios for pension funds, insurance companies, mutual funds, etc. And there's everything in between. From one end of the spectrum to the next, you have very different trading profiles, and thus are affected very differently by high-frequency trading.

For someone like myself, I make maybe a few dozen (relatively) small buys per year. These buys are usually in the neighborhood of 100 shares. If a high-frequency trading program jumps in and effectively front-runs me to to make a few pennies, I don't really care. Overpaying by a penny or two per share means nothing given my buy and hold (long term) strategy. I'm already out $9.99 per trade in commissions to my broker. I'm looking at a horizon of at least ten years, when these relatively small additional costs shouldn't matter.

On the other end of the spectrum is the big institutional investor, like the pension- or mutual-fund manager. This person's job is to constantly rebalance the portfolio to meet some pre-defined metrics; he's generally actively trading huge amounts on a daily basis. While he certainly wants to get the best price possible when he trades, it's practically impossible for him to do that given the volumes in which he deals. Unless he has a highly specialized trading algorithm---that is, something just as sophisticated as the high-frequency traders---he can't help but signal his intentions to the market. Telegraphing his intentions is what makes him a "victim" of the high-frequency traders.

I'm not a fund manager, but my assumption is that, like me and my small buy-and-hold strategy, he also doesn't care about having a small percentage skimmed off of each transaction. To me, it's like buying a big-ticket item, such as a car. Say you budget $27k to buy yourself a new car. Now, some enterprising company goes out and manages a massive, real-time database of every car available for sale in the country. This company can use this database to find you the exact car you want, right now for $27,250. If you're willing to spend $27k, do you really care if you pay an extra $250? And for that $250, you get precisely what you want, and don't have to wait. Compared to going to a dealer, who, if you're lucky, might have what you want at your price... but chances are, the dealer will have something close to what you want, and you'll have to negotiate the price. Or maybe the dealer can get you exactly what you want, but you'll have to wait while he works the intra-dealer process to provision the car. Or maybe he can get you exactly what you want, for even less than $27k, but you'll have to wait for the car to be manufactured. A car buyer can face all these scenarios, but I believe the fund manager most closely mimics the first: that is, he knows exactly what he wants, and he wants it right now.

My prediction is that we'll see the high-frequency trading landscape continue to evolve. Like anything, there will come a day when that kind of business and the skills required to do it are commoditized. And when it reaches that point, it will be much less lucrative. I think we'll see traders of all profiles using ideas and techniques from the high-frequency world in their own trading, meaning that the very people high-frequency traders take from will become direct competitors. The small-time trader like me will implicitly use such techniques, though they will be invisible, as it will actually be implemented by my discount broker (perhaps they'll offer

Re:short term skimming

[jonbryce]

Because high volume traders could use any changes in that 500ms to make arbitrage profits when they decide to say yes or no.

Re:short term skimming

[0100010001010011]

2 transactions a minute is 30,000 ms.

Re:short term skimming

[Anachragnome]

"The argument of 'providing liquidity' doesn't really seem like it has that much value for a normal investor."

Keyword is "normal". This sort of trading is anything but "normal". Only the largest entities can afford this sort of trading backbone--you and I have no such luxuries and have to rely on brokers to even stand a chance.

When I first scanned the summary I assumed that "Liquidation Center" simply meant an assload of processors to give them an edge in terms of transaction speeds/volumes so that they could "liquidate" soured stocks/commoditites/etc.

In short, they have invested in the ability to duck before the shit hits the fan, leaving everyone else standing directly in the path of said shit.

And another thing! With this sort of volume and speed, how in the hell does the SEC make sure it is all kosher? Are all of these trades being archived somehow, or does the SEC keep up with the trades? Or is this the point?--keep the SEC hopelessly back-logged that by the time they catch something it is far too late to really hold anyone accountable?

Re:short term skimming

[Anonymous Coward]

you've overlooked the fact that those fund managers very often work on behalf of the "small investor" who is trying to retire on a 401k or a pension plan.

Re:short term skimming

[pyite]

We paid for your dumb errors in the subprime crisis, so now that you are creating speculation on yet another imaginary value (physical closeness to the servers of the stock exchange ? Really ? Changes in the value of a company on the millisecond scale ? Are you serious ? ) you better show your whole scheme.

Whose dumb errors in the subprime crisis? Those errors were made by multiple parties. First and foremost, the people buying more than they could afford. It's taboo to demonize "Main Street," but let's face it, if people could, or chose to, do basic math, they would have realized they couldn't afford what they were buying. Second, stupid mortgage companies who relied on people's word that they could afford things. Third, those who thought tranching baskets of mortgages and pricing the tranches using default correlations for each tranche that were advantageous just to that tranche rather than rooted in reality.

These three groups of people have almost nothing in common with those who trade equities and equities derivatives. So, please, don't put two things you don't understand into the same bucket simply because both are products of "Wall Street."

Re:short term skimming

[LordNacho]

Finally, a sensible remark. It's really crazy how many people here on Slashdot hate finance more than they love technology.

Re:short term skimming

[Yvanhoe]

I have no problem considering a part of "main street" has some responsibility. But people who can't pay their debts is a problem known since millenniums (yes it is). Banker is a profession that is as old as this problem (yes it is). Having bankers unable to evaluate at least approximately the risk of a debt is like a farmer who forgot to plant seeds one whole year. Unforgettable. They showed grave incompetence, we now have to look behind their shoulders for their every move because no politician has the guts to make them pay.

Re:short term skimming

[LordNacho]

Good writing. You've explained why HFT is actually good for everyone, but only a very little bit. The piece you're missing is scalability. A good doctor can only save so may lives in a day, and makes an effort roughly in proportion to the number of patients. A broker can save a tiny few cents off a trade, but can do it for a huge number of clients, without extra effort. That's why they make more than doctors. Same goes for other almost pointless stuff, like sports stars, actors, authors, etc. Same effort, huge number of small beneficiaries.

Re:short term skimming

[Anonymous Coward]

While the average investor was partly to blame for not checking his actual buying power, remember that the banks' go to extraordinary lengths to hide the real cost of a loan so that a) home owners don't actually know how much they're going to end up paying in the long run and b) to make the loan look easier to pay than it actually is.

Don't cut the Wall Street mob any slack. They don't need, nor do they deserve it.

Re:short term skimming

[TubeSteak]

I'm not a fund manager, but my assumption is that, like me and my small buy-and-hold strategy, he also doesn't care about having a small percentage skimmed off of each transaction. To me, it's like buying a big-ticket item, such as a car. Say you budget $27k to buy yourself a new car. Now, some enterprising company goes out and manages a massive, real-time database of every car available for sale in the country. This company can use this database to find you the exact car you want, right now for $27,250. If you're willing to spend $27k, do you really care if you pay an extra $250? And for that $250, you get precisely what you want, and don't have to wait. Compared to going to a dealer, who, if you're lucky, might have what you want at your price... but chances are, the dealer will have something close to what you want, and you'll have to negotiate the price. Or maybe the dealer can get you exactly what you want, but you'll have to wait while he works the intra-dealer process to provision the car. Or maybe he can get you exactly what you want, for even less than $27k, but you'll have to wait for the car to be manufactured. A car buyer can face all these scenarios, but I believe the fund manager most closely mimics the first: that is, he knows exactly what he wants, and he wants it right now.

Talking about cars and manufacturing completely mischaracterizes the fluid and fungible nature of stocks.

The problem with your long winded analogy is that [enterprising company] and [fund manager] have access to the same database and the same stocks.
The only difference is a hundred mili/microseconds or so.

Whenever there's a technology revolution in any industry, we always see questions of "is it too soon to use this?" For example, vaccines in medicine.

No, not "for example, vaccines in medicine."
The first vaccine was created 214 years ago and we've had plenty of time to investigate and regulate.

High frequency trading, as we know it today, has barely been around for 5 years.
If you think that's long enough for the market players and the regulators to really understand the effects of HFT on the marketplace... well, not many people agree with you.

Re:short term skimming

[Antique Geekmeister]

It doessn't. Recent regulatory changes at the SEC, and promises to "review" the situation, have carefully avoided striking at the heart of the larger investment companies and groups that both engage in this sort of behavior, and have enough spare cash to fund politicians and arrange their own "reports" for the SEC on how disemboweling thihs sort of high-speed trading would "hinder the market".

It's insider trading with a high-tech cover. The data is not avaialble to ordinary investors in time for them to take keaningful action: the millisecond of having their servers right near the NySE overwhelm the reponse times of any ordinary trading entity and prevent their meaningful responses.

Re:short term skimming

[Anonymous Coward]

In fairness to the tranche modelers, all historical data indicated that foreclosures in geographically distinct areas were in fact largely uncorrelated. The housing bubble broke this assumption rather badly, but that was on those who made the bubble, not the quants.

Going back to the GP, if Joe Sixpack sticks to limit orders then millisecond trades don't affect him. Also keep in mind that it is not uncommon for brokerages to resolve trades in house rather than letting them reach the exchange for small or odd lot trades. For instance, most of my trades are on the $1000-5000 scale because I am just starting out. Most brokerages keep a shares of common stocks on hand so as not to pay the exchange to process the trades. They simply process the trade at the current market price and if their own reserves need adjusting, they can buy/sell in units of 1000 or 10000 shares.

Where High Frequency Trading really makes money is from trusts missing out on fractions of a percent - HFT is sort of a more legit version of the "steal the rounded off interest" scheme from Office Space - on an individual level it is meaningless, you may lose 1 cent per share, but doing this enough makes it profitable to the brokerages.

Re:short term skimming

[Dhalka226]

First and foremost, the people buying more than they could afford. It's taboo to demonize "Main Street," but let's face it, if people could, or chose to, do basic math, they would have realized they couldn't afford what they were buying.

While that's not untrue, it's somewhat disingenuous. If people knew basic auto repair, they could save a lot of money on oil changes and auto mechanics. If people knew basic home improvement skills, they could save a lot of money on handymen and repair guys.

But most people don't, and in that void of ignorance, fear or indifference exists entire industries who we collectively tag with the sometimes laughable title of "professional." A person doing some basic math might have figured out they couldn't afford what they were buying (and you're grossly oversimplifying the problem, by the way), but instead they relied on a series of "professionals" to do it for them -- professionals who are paid handsomely, not only in terms of their own salaries but in terms of commissions (real estate) and interest (banks/mortgage companies). Both of these parties nodded their heads emphatically and declared "of COURSE you can afford this, don't worry about it! Sign here!"

They did it from simple greed, and from a misguided belief that eh, even if we have to boot these freeloaders out of their house we have some of their money in pocket and real estate prices keep going up so we won't lose THAT much when we sell it to the next sucker. Through their greed, and their staggering unprofessionalism, they essentially collapsed two entire industries with a trickle-down effect that collapsed even more; industries that survive today only through the intervention of the federal government, right or wrong.

Some extra personal accountability is certainly a good idea to protect ourselves, but when we hire or deal with professionals I don't think any sane person expects them to be so wholly unprofessional as to tank their entire industry. They don't deserve to be let off the hook for that, not to any degree, even if peoples' ignorance is what allows them to operate that way. Ignorance, and more importantly knowledge of their own ignorance, is precisely why people hire professionals in the first place (not that there is much option when we're talking about a mortgage, I admit).

That these idiots are distinct from equities traders I do not deny, though I also don't think they're quite as separate as you make them out to be.

Re:short term skimming

[pyite]

In fairness to the tranche modelers, all historical data indicated that foreclosures in geographically distinct areas were in fact largely uncorrelated. The housing bubble broke this assumption rather badly, but that was on those who made the bubble, not the quants.

My point was a bit more subtle. Those holding equity tranches want defaults to be highly correlated. Although high default correlation means if one fails, they all fail, it also means if one doesn't fail, none of the others do either. So equity tranches were priced using that correlation. Senior tranches want little default correlation, because it means that defaults are random and will be absorbed by the equity tranches. Those tranches were priced using that correlation.

Where High Frequency Trading really makes money is from trusts missing out on fractions of a percent - HFT is sort of a more legit version of the "steal the rounded off interest" scheme from Office Space - on an individual level it is meaningless, you may lose 1 cent per share, but doing this enough makes it profitable to the brokerages.

Making fractions of a cent on spreads is a market maker's reward for providing liquidity and taking risk for a large price swing in the period of time in which they are still holding the securities (in the case of an underlier) or haven't yet hedged (in the case of some derivative).

Re:short term skimming

[AcidPenguin9873]

So only big banks are supposed to taking these risks and making huge returns, while Joe Taxpayer is not supposed to do that and just "live within his means"? Why should Joe Taxpayer be locked out of potentially big returns in a hot housing market? And why shouldn't Joe Taxpayer get a bailout when things inevitably go south? Banks take risks all the time to maximize their profits, and they were making huge profits (which turned into huge bonuses for executives) during the housing bubble. Joe Taxpayer was just trying to get in on the action.

Let me quote from this outstanding article by James Kwak:

But, let’s say I’m a guy who makes $15,000 a year. I realize, wow, I can get a $400,000 mortgage and I can live in this house for a few years, and if housing prices go up, I can flip it and I can actually make a couple hundred thousand dollars. And let’s say I’m really clever, and I say, if housing prices go down, I’ll just walk away and I will have gotten to live in a really nice house for three years at no cost to myself. I mean, that’s the worst, most cynical spin you can put on it, right? But this is exactly what people on Wall Street do. The person who is criticizing the janitor for doing this is the same person who thinks that businesses should exploit every legal opportunity to make profits. So even if you attribute the worst possible state of mind to the guy making $15,000, he’s still just doing what any businessman should do under the circumstances. But our national ideology somehow doesn’t allow us to think about it in those terms.

Re:short term skimming

[sjames]

The first mistake main street made was trusting a single word that came out of lender's lying mouths. They were talked in to believing they could afford it by "experts" who "knew what they were talking about".

As for others on Wall Street, they're cut from the same cloth. They use the same screwy math and the same tired excuses for amoral behavior. They occupy the same buildings and run in the same circles. The same people pull the strings. When someone shoplifts, we don't punish that evil thieving right hand, we punish the person.

Re:short term skimming

[totally bogus dude]

High frequency trading, as we know it today, has barely been around for 5 years.
If you think that's long enough for the market players and the regulators to really understand the effects of HFT on the marketplace... well, not many people agree with you.

I think you missed the point of the analogy. Our daily lives are full of things that were once newfangled technology that many people resisted and demonized, saying it would destroy this-or-that aspect of society. Most technological advances don't destroy all that much, and many have provided enormous improvements in our quality of life.

HFT may be new, but people will certainly be studying it to see what effects it's having, and incremental adjustments to it will likely be made to try to maximise its benefits while minimizing its drawbacks. While I tend to agree that new technology tends to improve things, it's also worth remembering that most new techs also bring new problems. So just because HFT may have some negative impact on the market, that doesn't mean it's not overall an improvement. Just like the introduction of the automobile saved us from the pollution and sanitation problems of horses, but brought with it problems in the form of exhaust gasses. And the electric car will solve that problem but create others, and so on...

It's not just a black and white "this new tech is good" or "this new tech is bad" - every time we solve one problem, we usually create another, but that's the very nature of progress.

Re:short term skimming

[OeLeWaPpErKe]

You'd think, indeed, that programmers should know that you simply cannot solve a program's bugs by adding ever more code (in this case the program is the law). The subprime crisis was created by people following the rules, knowing they had a loophole that allowed them to throw all the risk on a few institutions (that had received guarantees of government backing). Okay, this is probably a bug in the law, so what do we do ? We add a few more rules hoping they make it impossible to "hit" the bug again. Good luck with that.

Why, in the name of the almighty atheismo, aren't we cutting out the hole in the law in the first place (ie. dissolve freddie and fannie entirely and find another way of doing this)

Or at least, you'd think programmers would realize that if a program doesn't fit in a gigabyte of source code, it's HIGH TIME to cut out large parts of what can't be anything but garbage.

Re:short term skimming

[budgenator]

I'm sure this has been fixed because an Average Joe can easily make serious cash off a flashcrash.

Re:short term skimming

[Builder]

banks' go to extraordinary lengths to hide the real cost of a loan

Really? I hear this a lot, and I can only assume that this is an american thing. When I got my mortgage, it was fixed at a specific rate for 3 years. After that rate, it would track the base rate as set by the bank of England + 2% for the remainder of the loan.

At the same time, all initial fees were itemised and all potential future fees including late payment and final settlement fees were described.

Who would sign a contract _without_ knowing all of this in advance?

Re:short term skimming

[Coren22]

Mostly people who can't read. This is the same here in the US. I was given all the terms of the loan right up front.

Re:short term skimming

[Coren22]

When I was looking for my first house, my parents gave me a good rule to follow. The most expensive house you should buy would be three times your yearly salary. When I got my credit run by the bank they laughed at me looking at such a low number and suggested another 2x that, I ignored them, and thank god I did. If I had not stuck by that rule, now that I make twice what I was making then, I could not afford the houses they were trying to push me for. I think it all boils down to personal greed, the bank and real estate agent make a percentage of the sale, so they will push you for the most expensive house you can "afford".

Re:short term skimming

[Coren22]

Those numbers are a bit suspect. There is no way a guy making 15k could afford a 400k loan even for the first couple months. at 2.9% interest (a typical subprime at the time) and a 30 yr mortgage, that is 1664.92 a month. That is without PMI, which most people would be forced to pay, and no property taxes. 15k is 1250 a month, about $875 takehome (taxes are about 30% around here, might be even less on this little money). I suppose if it was interest only it is possible, but that is just asking for trouble, and I can't imagine getting a rate low enough to afford the payments still. At the above rate, the interest only payment is 966.67.

Re:short term skimming

[BStroms]

Only the largest entities can afford this sort of trading backbone--you and I have no such luxuries and have to rely on brokers to even stand a chance.

If you want to day trade or do other short term trading. However, that was almost never a good idea, even before high-frequency trading. It's still possible for the average investor to take more of a Warren Buffet approach to investing. Always go for the long term. He's advised people to think about stocks as owning a stake in the company and to invest in ones you believe have a solid long term business strategy. Considering yourself a partial owner makes you feel more tied to your investment and less likely to reflexively sell because of short term fluctuations.

Re:short term skimming

[Builder]

I'm not really sure that the rate of stupid or illiterate people in a country is the fault of the bank :(

Re:short term skimming

[Coren22]

The illiterate was kind of a joke, but it mostly is people who just don't bother to read the terms, or are simply unable to understand them and are not explained the terms thoroughly enough.

Re:short term skimming

[Decker-Mage]

You missed a group there. The politicians had been pushing on Wall Street for decades (since 1968) to price down home mortgages (drive down interest rates reflected by 'red-lining' where it was rational behavior) via the Fed and various pieces of legislation (Safe Housing Act, among others). Wall Street reacted by coming up with methods to effectively reinsure those mortgages using CDO's and any other method they could think up over that same period. For it to work, Wall Street had to have to cooperation of the Fed as well as the politicians and for several decades it did. There were several near collapses averted over the years but the Fed successfully stopped system-wide revalueation. Finally, when it did start going off a cliff, some people and institutions were left still looking for a seat when the music ended (e.g. J. P. Morgan in investment banking, too many to count in consumer) but didn't have some sugar-daddies waiting to pull them out of the wreckage (GMA, Goldman-Sachs, ...).

The sad thing about this whole mess is that the CDO's and other instruments still have 90% of the assets still performing but in a gift from Congress and the Fed the institutions were able to write-off the entire asset, while still keeping the stream of payments. As a result you'll be seeing banks and other investment firms having record future profits. Nice.

Re:short term skimming

[Anonymous Coward]

Anyone got an address on these? Perhaps we could run an ad in the personals section of the next edition of Inspire.

Re:short term skimming

[Red Flayer]

Who would sign a contract _without_ knowing all of this in advance?

That's not the problem. The real problem is the people who knew this in advance, but were unable to understand the implications.

I bought a home in 2003 -- my first house, as I'd finally decided to move to the burbs with my wife & start raising a family. I worked with three different mortgage brokers until I found the best mortgage for us.

Here are examples of some of the things I was told:

Don't worry about the interest rate after the first five years on this 5-1 ARM. You'll be selling at a HUGE profit before then.

You should buy the biggest house you can afford so you can maximize your profits when you sell. We calculate you can afford to spend 40% of your take-home on your mortgage, so we're willing to approve you for $X. 40% is pretty typical these days for people of your income.

If you get this 3-1 ARM, your monthly payment will only be $Y if you purchase a home of cost $Z. Let's go over your monthly budget so we can see how to get you into the house of your dreams with this mortgage.

The reason I give these examples is to show how easily people can be misled by mortgage brokers, how easy it is to get people to believe what they want to believe -- that they can afford a house and a mortgage that are both too much for them. Never mind the fact that so many people were hoodwinked into believing that the housing market would continue to boom forever.

So yeah, I read all the terms and decided on a 15-year fixed mortgage. But I think people like us are in the minority -- most people just don't have the math and logic skills necessary to solve this problem on their own.

Re:short term skimming

[maxume]

The mortgage companies were more than stupid, they were corrupt. And they were writing bad mortgages so they could sell them, in many cases they weren't missing the lies being told by loan seekers, they were encouraging the lies.

Re:short term skimming

[purpledinoz]

Your car analogy does not work. High frequency traders don't help sellers match buyers. They skim a little bit off of sellers and buyers. They are distorting the market in a major way. A market functions well when everyone has the same information to make their decisions. High frequency traders are given information not available to everyone (they get the information before everyone else gets it). I do understand your vision of HFT eventually evening out through competition and being the norm, but I also see the vision of the 4 big banks gaining a permanent advantage over everyone else.

in-equity

[alphatel]

Not only are you completely powerless to do anything about it now, but when some glitch causes your pension fund to suddenly be worth 10 cents, you won't be able to sue anyone.

Re:in-equity

[SirGeek]

Or someone manipulates the markets and causes an mini-crash again.

Re:in-equity

[Anonymous Coward]

Or you could wait 72 hours.

Re:in-equity

[Anonymous Coward]

Not only are you completely powerless to do anything about it now, but when some glitch causes your pension fund to suddenly be worth 10 cents, you won't be able to sue anyone.

Yeah, but I'll still be able to shoot someone. Like a banker. Does it matter to me the banker I might shoot isn't the banker that robbed me? Nope. Steal all my money when I'm too old to work, and I'll sign up for one of those places that gives you free room/board; possibly a nice injection to help me sleep forever.

Re:in-equity

[SuricouRaven]

It wouldn't even need deliberate manipulation. When there are thousands of programs all making the same decisions on the same input, even a natural fluctuation can trigger a disaster. A stock falls a bit, programs see it, and within milliseconds are selling - triggering a further fall, a feedback cycle of collapse.

Re:in-equity

[biryokumaru]

This kind of thing always makes me wonder why you see so many homeless people. They could just smash someone's head in with a rock and have a nice, clean, warm home with three squares a day and plenty of time to read or watch TV.

The reason is, those homeless people, unlike the bankers, aren't heartless sociopaths. I think when the bankers rob you of everything you've spent your life saving, you may find the same is true of you, unfortunately.

Re:in-equity

[Pinky's Brain]

That's 72 hours of liquidity gone, there is an opportunity cost there ... an opportunity cost measured in true dollars going into the pockets of speculators. Basically we have traded wider spreads for higher instability, is it a good trade? Maybe, dunno.

Personally I think bids and offers should be matched up only once every hour ... I don't see any need for this stuff to happen at wire speeds.

Re:in-equity

[Surt]

One would hope, though it would not always be true, that there would be an equal number of programs seeing an unnatural downward deflection and issuing buy orders.

Re:in-equity

[wowbagger]

This kind of thing always makes me wonder why you see so many homeless people. They could just smash someone's head in with a rock and have a nice, clean, warm home with three squares a day and plenty of time to read or watch TV.

Methinks you play The Sims too much - or is the homeless person who's clunked me on the head not going to have to explain himself to my friends who find him in my home and me nowhere to be found?

Or are you implying that prison is better than homelessness? That the threat of larger, nastier people, who might seek to stick a shank (or worse) into you for fun, that the loss of freedom, is better than homelessness? Perhaps it is not a lack of being a sociopath but rather the simple calculation that even a sociopath can make - that freedom is better than imprisonment.

Re:in-equity

[SuricouRaven]

Where would the profit be in buying a declining stock?

I oversimplified the example for clarity - in realise the algolrythms used are far more complex, and include some safeguards against this type of collective behavior, otherwise it would happen all the time. Still, the possibility is there. The 2010 Dow mini-crash was caused in large part by high-frequency tradeing which amplified a minor variation into a severe one - and came dangerously close to progressing into a full-blown crash. It was only a combination of a final emergency safeguard closing the market for fice seconds and a lot of blind luck that averted disaster. High-frequency tradeing, espicially of the extremes seen today, is just inviting another such incident - and with the frequency of tradeing constantly going up, next time there might be no stopping a crash until it's too late.

Re:in-equity

[SuricouRaven]

And before anyone mocks me for that absolutly terrible spelling error: I'm short on sleep, caffeine, and... I can't even be bothered to finish that sentence. Consider my excuse preemptively made.

Re:in-equity

[Sponge Bath]

"...nice, clean, warm home with three squares a day and plenty of time to read or watch TV."

Prison?

I've met plenty (certainly not all) of homeless who are heartless sociopaths. Some of them are heartless psychopaths. Don't let your compassion cloud your view, there are some scary people on the streets.

Be suspicious of those you don't know well, no matter how they dress or where they work, and avoid being bitter when they do bad things to you.

Re:in-equity

[Anonymous Coward]

If you match bids and offers once per hour, you're going to widen the spread pretty dramatically. It's no coincidence that the spreads have tightened up as trading speed has increased. If you force market markers to hold securities for at least one hour, they're going to have to widen their spreads pretty dramatically in order to compensate for the risk.

You would also see significantly higher volatility in the market. The way things currently stand, you can get a very good idea of what a security is worth at any given instant. If you have to wait an hour before the market updates, it's going to be hard to predict what the next trade price will be. That's going to result in a lot more price volatility, which will also increase spreads.

Another consequence is that you're likely to put smaller shops completely out of business. Holding a security for a minimum of one hour in an environment where pricing information is not available is just not something that smaller firms will be able to do (if nothing else, net capital regulations will foreclose it, due to the enormous risk). For the big firms, it will be like a return to the good old days- all those pesky HTFs gone, and only 2-3 market makers for any given security. This, more than anything else, will increase spreads and spell the end of the exchanges and a return us to the days of market makers. "You want to buy 100 shares of MSFT? You'll pay what we say. What are you going to do, go buy it from someone else? Good luck with that!" "You're ready to sell your 100 shares of MSFT? You'll get what we're willing to give you. What are you going to do, go to someone else? Try to trade it on an exchange? Har har har!"

This is a complex situation. The rise of the HFTs was one of the biggest shakeups in Wall Street history- the large firms lost huge chunks of their control of the market, and are still reeling from the shock of lost profits. They would be more than happy to see the end of HFT and a return to the days when they controlled the market. HFT's may cost the market $0.01/share, but that's nothing compared to the bad old days of monopoly market makers extracting $2/share or more.

Re:in-equity

[Anonymous Coward]

You mean bids and offers should be matched 24 times per day?

Or my bid and offer each last 60 minutes?

If you're not careful, don't you just end up with things going at wire speeds, just with more rules?

Re:in-equity

[Anonymous Coward]

The liquidity is still there, it just may not be a price you agree with. If you wanted to sell at a predetermined price at any time, you should have bought some options.

Re:in-equity

[jonbryce]

Value trading. You buy it in the hope that it will bounce back again. For example, the people who bought BP shares while the oil was spilling out onto the Gulf of Mexico are now sitting on large profits.

Re:in-equity

[Estanislao Mart�nez]

Where would the profit be in buying a declining stock?

It's called short selling stock.

I think you're missing an important detail here.

Re:in-equity

[Gorobei]

Personally I think bids and offers should be matched up only once every hour ... I don't see any need for this stuff to happen at wire speeds.

Numerous markets had, and do have, this "feature." For example, trade-at-close aggregates all the orders at the day's close, and everyone gets the same price if the close price is within their limit.

The effect of this feature was to benefit the technologically sophisticated traders (I was one back in the nineties.) You waited until a few seconds before the close, then slammed limit orders onto everything that moved: the poor retail guy who had put in an order 1 minute ago was your counterparty, and he had just given you a free option for 58 seconds.

Re:in-equity

[Anonymous Coward]

What about this 10% rule that is now in place? They say in that documentary the "circuit breaker" has been tripped several times since may 2010. Is it enough of a safeguard?

Re:in-equity

[LordNacho]

Yes, this is a real issue. But consider what happened in the mini-crash. Inside of an hour or so, everything was back to normal. If old school flesh and bone market makers had seen the market collapse hundreds of points in an hour, they would NOT be jumping back in immediately. They'd probably have at least gotten themselves out and then closed the day at a big minus. This then would cause waves in other markets, people would start talking about what to do overnight, and there would be all sorts of spurious explanations of why the market had to collapse, which would all feed on itself. Perhaps, as they read that not much had happened as a real cause, the market could come back to where is was, but more people would have been affected.

Re:in-equity

[LordNacho]

And what's wrong with speculating? We all do it in one way or another. Taking a degree in the hope of getting a job. Buying a house. Holding cash instead of hard assets. Changing jobs. Speed dating. Many of life's decisions are made in the hope of an uncertain gain.

Re:in-equity

[Nimey]

An hour's a pretty long time, these days. I could see once a minute, or once every five minutes.

Re:in-equity

[Anonymous Coward]

They are matched up with 100-200ms delay. The "big boys" pay extra to get trade data *before* it happens, by about 50ms.

And no, I'm not posting this Anonymously because it's a trade secret. This is public knowledge and exchanges are even advertising this type of fraud. HFT is where the Joe Blow is a loser every time. You have a spread

    Call: $11
    Put: $10

then HFT come in between the two and buys at $10 and sells at $11. In old days, the $10 trade would go trough to the buyer for $10. Now, HFT computers skim off any middle ground. Of course this is not as simple as I illustrate, but HFT are setup to see the market *in the future* by 50ms, hence they always win..

There are other ways to make market. The old "buying all until price X" (limit) today means you'll buy ALL at price X. More explained here,

http://seekingalpha.com/article/151173-hft-the-high-frequency-trading-scam

Re:in-equity

[Surt]

The profit in buying a declining stock comes in catching the rebound.

Re:in-equity

[Anonymous Coward]

"Or are you implying that prison is better than homelessness?"

He's implying that if the social safety net homelessness becomes worse than prison, get used to checking your six.

A starving man trapped in a blizzard will absolutely cave your face in if that's his only path to a cup of soup and a place to get out of the wind. He doesn't have to be a sociopath at all.

Re:in-equity

[Anonymous Coward]

The profit would be in the classic idea of buy low-sell high, i.e. buy on the dips. Granted, any complex model would also examine higher order derivatives, etc, but it makes sense to buy on small declines.

Re:in-equity

[Steeltoe]

Well, the point is that you never actually know wether a stock will rise or fall, so what do you do?

If you go chasing after "rising stock", you will hit the top more often than not, and thus actually lose money pretty fast when fees are deducted.

So real traders always buy a declining stock in a positive trend. They time their trade so that they know it should rebound and continue the positive trend, preferably more often than not to be profitable. The excellent traders, can trade sideways and falling markets as well. The quants buy intercorrelated markets, HFT and a bunch of other neat tricks. There are many ways for those who investigate.

Since traders trade in all kinds of timeframes, there's always someone interested in buying something which is falling on another timeframe, if there is interest at all in that paper. To imply that this is not always happening, seems a bit ridiculous to me, but it seems people always want to rationalize away what they can't accept. However in extreme situations, the market can be overloaded by a cycle and "crash", even when the cycle is very short-term, or it can aggravate other financial troubles as well.

Extremely sudden drop-and-rally or rally-and-drop in a positive trend, are quite common in normal markets, and may signal overall direction of the market or reversal and first target, with the reverse being likely for a negative trend. When doing technical analysis, they are frequently occuring in all timeframes, and are not a "technical glitch" at all, but rather the synchronicity of different cycles opening a window into a probable future of the price. Often the "glitch"'s reversal price coincide almost perfectly with some MA, unless stopped by the "overseers" of course. The MA can tell you something about the period of the cycle and give you time to prepare.

This "signal" is not easily tradable though. When you discover it _reliably_ as a reversal signal, it is often too late. It can also be a very bullish signal at the start-middle of a cycle, shaking off many would-be investors with too tight stops. However, it can be used as a hint, and a warning of what to expect, or in conjunction with other signals. A trained eye or neural net should be able to translate its meaning more reliably than the most simple rules.

Since a market will not move unless there is both buying and selling, it doesn't make sense to me that the move made by the "mini-crash" can be explained away as a technical glitch. The buying and selling system was working, it was just that people were not prepared to accept the results. If true, it speaks of a terrible mother of all bubbles to be burst in a few years.. We won't know exactly when or how, since it was stopped mid-way. Since then the markets has been manipulated in big ways in order to expand the bubbles even more, making it hard to estimate. My guess is 2-3 years from now, or even sooner. For now, markets are still bullish, probably until April / May, and then we may hit some really volatile periods.

What also make markets work, is that not everyone is an expert, and thus loses money to the market. Those who make the quickest profits, are also often the ones who go bust in the most sudden and "unexpected" ways! So you can't really assess someone on their equity curve, even when not faked, since it says nothing about risk..

We must never forget that the markets are not infallible, and they cannot and should not make the decisions for us.

Re:in-equity

[vux984]

Trivially solvable by matching orders at random intervals.

But honestly, it doesn't need to be hourly. Match them once a second, and limit you to transacting a particular stock more than once a minute.

Re:in-equity

[umghhh]

I do not mind much HFT albeit I question its usefulness for the society and economy at large. HFT is pointless but it is legal and it may stay so. The problem lies elsewhere actually - if you look at the system in general - it does not have built in controls at all - the only thing that prevents it from violent movements is its mass. It works well most of the time providing monies to HFT masters but when all goes wrong then we are all sorry and guess who is going to salvage the financial institutions?

I guess some sort of negative feedback loop needs to be built in - transaction tax would help I think this run to microseconds while providing money that could be used if financial systems crash which they do from time to time anyway. I think this is undoable though as this would mean that rules setters would refuse money from the HFT masters which is unlikely.

Re:in-equity

[umghhh]

Well let them fix that period at which to trade alone - introduce transaction tax (small but for any transaction on financial market) and this run for microseconds will cease.

You will not see it happening any day soon of course for obvious reasons.

Re:in-equity

[Pinky's Brain]

You can't look at volatility in isolation ... the risk of selling just before the cut off in a flash crash represents a risk in the current situation which can't be accounted for with the nice steady state low spreads.

The risk of corruption and market manipulation (which we know for a fact has already occurred, see the whole information dealing controversy with the exchanges) will cause an across the board depreciation of all stocks not captured in the lower spreads.

The costs of HFT are very hard to calculate ...

Market makers have no easier time determining the direction the market will go as anyone else. If the market makers try to extort a consistent high spread in a once every hour match up there will consistently be money to be made every hour by undercutting them. Of course this assuming they aren't given privilege, which they shouldn't.

Re:in-equity

[Pinky's Brain]

I don't believe in giving any market participants privilege either, everyone should be able to directly put their bids in without leaking information to brokers or market makers.

Re:in-equity

[maxume]

The flesh and bone market makers noted the glitch and temporarily halted floor trading. The reversed trades were all on electronic markets.

I Can't Help But Think...

[sycodon]

...that automated trading has and will cause more trouble than it is worth to the overall economy.

Re:I Can't Help But Think...

[Anonymous Coward]

Well, it is a silicon arms race after all. Can't wait till they through a little AI in there to help optimize trading along the way. Who knows. It might become self-aware and flip us all the finger.

Re:I Can't Help But Think...

[biryokumaru]

"That's funny... all our HFT algorithms keep investing in our own datacenter technologies..."

Re:I Can't Help But Think...

[Prof.Phreak]

...wish I had mod points :-)

Whoop De Doo

[Iphtashu Fitz]

Looks like virtually any other commercial datacenter I've been in. Nothing I saw these articles leads me to believe they're any different. Replace "Wall Street" with virtually any other company with an internet presence and you get the same thing.

Re:Whoop De Doo

[Anonymous Coward]

You'd think they would at least put some bling on for the photos. Golden racks, a diamond logo or something. Google image search for "datacenter hallway" returns hundreds of photos that look the same.

Re:Whoop De Doo

[Anonymous Coward]

Mahwah is a state of the art DC. Has bomb prof walls and roof; power is supplied by 2 different power grids from different states. Isolated pods that act as stand alone datacenters within the DC. There is also infrastructure built around it to prevent car bomb attacks.

Re:Whoop De Doo

[ColdWetDog]

The blue lights. According to TFA,

And yes, there are blue lights to keep things cool – both the equipment and the visuals.

I'll bet your clunky ol data center doesn't have lots of blue lights.

Re:Whoop De Doo

[SuricouRaven]

There is just one difference, though. The trading machines, regardless of their location, have the same length cable to the switch. Even if it means coiling some of it up. The latency demands are so strict, the customers even care about the cable length - and it's just easier to give all the customers the same length than to maintain tiered pricing on the racks.

Re:Whoop De Doo

[PolygamousRanchKid]

You'd think they would at least put some bling on for the photos. Golden racks, a diamond logo or something.

Monster Cables. They make you think that your quality of Ethernet is better, because they are expensive. Expensive stuff is better, right?

Re:Whoop De Doo

[nullifi]

We installed a storage array that has bright (flashlight level) blue lights. They drive me mad, and I would put tape over them if my boss would let me.

Re:Whoop De Doo

[biryokumaru]

No, no, it's wall to wall Denon AK-DL1.

Re:Whoop De Doo

[Anonymous Coward]

Ah, so that's how I'd encourage employees I don't like to quit. I like it. Subtle, yet invasive.

Re:Whoop De Doo

[Spectre]

No, no, it's wall to wall Denon AK-DL1.

That's an April Fool's Day "joke page", surely?
Please tell me there aren't people so insane as to believe an error-corrected digital transfer of data is going to sound any different no matter what cable is used between devices, until it is so bad that it can't keep up with the data stream at all ...

Re:Whoop De Doo

[vlm]

There is just one difference, though. The trading machines, regardless of their location, have the same length cable to the switch. Even if it means coiling some of it up. The latency demands are so strict, the customers even care about the cable length - and it's just easier to give all the customers the same length than to maintain tiered pricing on the racks.

Its marketing. The propagation delay between different pairs can vary by up to 50 ns due to the different twists. That is why the fancier "VGA over CAT-5" converter box thingies have a skew compensator. If you use a layer 1 that uses all 4 pairs it doesnt matter, but if you use a layer 1 that uses only two pair, then theoretically some customers will have a ping time 50ns lower than the slowest customers. one foot roughly equaling one nanosecond means the electrical delay can't be specified more accurately than 50 feet. Noobs have all kinds of fun with TDRs because of this.

You'd be way the heck better off using fiber, as first of all, it will actually work, and secondly, its more expensive so it must be better (aka monster cable marketing)

Re:Whoop De Doo

[SuricouRaven]

It actually would be lower latency too. The speed of light in glass is higher than the speed of electrical signals in cat5. I don't know the cat5 speed from memory, but IIRC the speed in coax is around 2/3 C.

From what I've read in the networking magazine we get at work, the current debate in high-speed tradeing IT is 10-gig ethernet vs infiniband, with the latter offering lower latency in the end devices.

Re:Whoop De Doo

[drsmithy]

It's absolutely real. Check out the Amazon page for some thoroughly entertaining reviews.

Re:Whoop De Doo

[Anonymous Coward]

Subtly take a black sharpie to the lights. They still shine through, but not nearly as bright.

- Pitabred, anon because I've modded

Re:Whoop De Doo

[Anonymous Coward]

Car bombs? The attack MUST be nuclear. Anything less would have no meaning to the enemy. The safe placing of data centers from NYC has been going on for years. There is a section of Lyndhurst, NJ called "Wall Street West" just south of the SR 17 spaghetti junction with SR 3 where it becomes Ridge Road. Just a mere mile north is the Federal Reserve Bank of New York compound built on land acquired from Becton-Dickinson. I would be reasonable to assume that the maximum underground depth of said structures would be classified. Who here would be foolhardy enough to drive by on SR 17 south with a van load of ionizing smoke detectors?

This sounds like a job for someone would enjoy rotting in ADX Florence for the rest of their [radiation shortened] days...

Re:Whoop De Doo

[Anonymous Coward]

Mahwah is also quipped with fire suppressant that wile stops fires does not short the equipment.

EMP?

[MikeKD]

Hopefully they're hardened against EMPs.

Guidos installing pizza boxes in place of blades

[DanCentury]

Jersey? They're too close to Snookie for my comfort. The end is near.

If this was Fark (and 2003) I would be making a photoshop of guidos installing pizza boxes in place of blade servers.

So, where are the VAXen?

[Tackhead]

Oh, that's right. Even 22 years later, VAXen, my children, just don't belong in some places :)

Re:So, where are the VAXen?

[Anonymous Coward]

Was that you?

Re:So, where are the VAXen?

[Anonymous Coward]

I wish, but thanks for the implied compliment :)

the heart of wall street is the bond markets

[Anonymous Coward]

not the stock market.

that whole 'shadow banking system', you know, that like, crashed and stuff, that we are paying for with 2 trillion dollars. most of that didn't go to buy stocks.

Re:the heart of wall street is the bond markets

[Anonymous Coward]

Thank god, atleast someone understands this!!

Humans in the loop.

[blair1q]

“Markets are there for capital formation and long-term investment, not for gaming,” [Michael Durbin] says here

Amen to that. The markets should operate as though there are humans at every step. Otherwise there's no need for humans on the edges, either.

Re:Humans in the loop.

[Anonymous Coward]

The markets should operate as though there are humans at every step. Otherwise there's no need for humans on the edges, either.

Um... Without humans who want to benefit from trade, there is no need for markets at all.

Re:Humans in the loop.

[blair1q]

Convince an HFT program of that.

Re:Humans in the loop.

[blair1q]

In other words, computers are perfectly capable of putting every stock on the market into the rail, and it's the human traders who are keeping market value anywhere within praying distance of the actual value of the thing being traded.

Re:Humans in the loop.

[Anonymous Coward]

And as soon as the machines figure that out, cue Skynet.

Re:Humans in the loop.

[Palpatine_li]

Then you'll have less liquidity, which is the exact cause of the (first) Great Depression.

Re:Humans in the loop.

[khallow]

Then you'll have less liquidity, which is the exact cause of the (first) Great Depression.

So low liquidity caused firms to be valued at dozens to hundreds of times any rational value? Low liquidity passed the Smoot-Hawley act? Low liquidity forced FDR to pass legislation and form oligopolies favorable to his labor union and business backers, to steal the wealth of tens of millions of Americans by seizing their gold, and to attempt to pack the US Supreme Court in an attempt to bypass the US Constitution?

No offense, but I see deflation/low liquidity as an effect not a cause. People hoard money and other assets, and don't trade on the stock market when they've just been burned badly by a deceptive stock market, fraudulent banks, and rogue governments.

Re:Humans in the loop.

[khallow]

Amen to that. The markets should operate as though there are humans at every step. Otherwise there's no need for humans on the edges, either.

Conversely, if I want to use a market, I don't want it saddled with Luddite hysteria. The value of a market is not in the employment it provides to us, but to the goods and services we can acquire through it. I couldn't care in the least, if there is a complex ecosystem of unsupervised trade programs operating at minute time scales on the market. Instead, like any tool, I merely wish it to work when I use it.

Re:Humans in the loop.

[blair1q]

When the machines are making 10,000 trades a second, you have no way at all ever to complain about a lack of liquidity.

Re:Humans in the loop.

[blair1q]

And, as usual, "I merely wish it to work when I use it" ignores the effects of the mechanization.

A system made from human beings can* adapt to your requirements for "just work", while machines will force you to conform to their conventions.

In this case, the machines are not making an efficient market. They are trading according to their own rules, far more often than the human who created those rules could do it himself, and with no reflection at each iteration as to whether it's the right thing to do given the outcome of the last iteration. When several such machines are operating at the same time under differing sets of canned rules, the "market" no longer reflects human valuation but a mechanical equilibrium. When you enter your trade, it is not at the value of the item but at the value to which the enumeration of that price has been pulled by the system dynamics of the computers that are ruling the input streams. I.e., when you sell into a computer-driven flash-crash, you're getting fucked by a mechanism, not participating in a market.

That's the opposite of what markets are supposed to be.

* - not that they all do, but this is because of management policies that make the worker-humans work like machines, according to preset rules that make no sense when user-human expectations are pushed into the input hopper.

Re:Humans in the loop.

[khallow]

And, as usual, "I merely wish it to work when I use it" ignores the effects of the mechanization.

That is by intent.

A system made from human beings can* adapt to your requirements for "just work", while machines will force you to conform to their conventions.

Your footnote for the "human being" choice kills your argument.

* - not that they all do, but this is because of management policies that make the worker-humans work like machines, according to preset rules that make no sense when user-human expectations are pushed into the input hopper.

In other words, the only time we'd be comparing machine and human systems, the human system will be constrained. I might add that these circumstances will hold for market trading systems.

To be blunt, for my purposes, the only difference between a human run stock market and a machine run stock market is that the human one will be slightly slower, much more error-prone, have a larger price increment, and be a huge pain to use since I have to interact with a human rather than just make my trade by machine. That's not interesting to me.

In this case, the machines are not making an efficient market. They are trading according to their own rules, far more often than the human who created those rules could do it himself, and with no reflection at each iteration as to whether it's the right thing to do given the outcome of the last iteration. When several such machines are operating at the same time under differing sets of canned rules, the "market" no longer reflects human valuation but a mechanical equilibrium.

They experience the same dynamical forces as a human trader. There will be the same tendency towards an efficient market.

When you enter your trade, it is not at the value of the item but at the value to which the enumeration of that price has been pulled by the system dynamics of the computers that are ruling the input streams. I.e., when you sell into a computer-driven flash-crash, you're getting fucked by a mechanism, not participating in a market.

And the solutions are: a) don't trade on margin, b) don't make market trades, and b) don't place stop orders. Those three choices fix the problem mentioned above.

Re:Humans in the loop.

[blair1q]

I'm not talking about taking machines out of the process of making your transaction easier.

I'm talking about taking machines out of the process of deciding to make transactions.

And no machine can "experience the same dynamical forces as a human trader". Human beings have insight outside the limits of the program the machine uses. No machine can do anything volitional until a human being tells it to. So the machine is just something doing the same thing a billion times, instead of a human making a human decision a billion times.

As for your "solutions": a) margin is irrelevant, and should be avoided even in human-only markets if you're skittish about amplifying risk; b) if you never make market trades you separate yourself from immediacy, which is an exploitable inefficiency in your strategy; c) is in conflict with b, since a stop is just a limit trade which is the opposite of a market trade so if you shouldn't do market sales or limit sales you'll never sell, ever. None of these speak to the central problem, which is that the idea that machine-driven trading creates liquidity and improves the market's accuracy is a fallacy. Machine-driven trading is just as likely to create scarcity and more likely to create false attractors for the price, creating inaccurate valuations for the thing on which the trading vehicle is based. The idea that making trading easier is more important than determining valuation accurately is also a fallacy. But since trading shares generates commissions, and holding onto shares for value does not, the people whose income is based on commissions will always argue for more trading. It's up to the people who want things valued accurately to tell them to shut up and do the jobs they have properly instead of cadging for more busy-work with no productive value.

Re:Humans in the loop.

[khallow]

I'm talking about taking machines out of the process of deciding to make transactions.

Again, this is why I don't care. My transactions are all made by me.

And no machine can "experience the same dynamical forces as a human trader". Human beings have insight outside the limits of the program the machine uses. No machine can do anything volitional until a human being tells it to. So the machine is just something doing the same thing a billion times, instead of a human making a human decision a billion times.

The dynamical forces are an inherent property of the market not of the traders. So yes, the machine experiences the same dynamical forces as the human trader. It's like claiming the Earth would move differently through space if it were sentient. But Newtonian mechanics doesn't change, if you are sentient or not.

Further, a machine doesn't need to have human flexibility for a lot of market related tasks. Some are beyond the capabilities of a human. If you want to make a billion trade decisions quickly, then you need a machine. A human won't cut it.

As for your "solutions": a) margin is irrelevant, and should be avoided even in human-only markets if you're skittish about amplifying risk; b) if you never make market trades you separate yourself from immediacy, which is an exploitable inefficiency in your strategy; c) is in conflict with b, since a stop is just a limit trade which is the opposite of a market trade so if you shouldn't do market sales or limit sales you'll never sell, ever.

The thing all three of these have in common is that you don't have complete control over the process. Margin trading opens you up to liquidation risk where your assets are sold off, without your consent, to maintain a certain level of reserves in the account. Market trades trade at whatever the market spread is, not what you thought the spread was when you initiated the trade. So yes, you could end up selling on the bottom of some computer crash without your knowledge (at least until you get updated market information on your trade).

Finally, by "stop" orders I meant "stop loss" orders, which are market orders which trigger only when some deviation from the trade price occurs. The traditional use of these is to sell out of the security, if the price falls too much.

What you thought I meant by stop order is a "limit" order. For these, you have complete control. You can't possibly trade beyond the limit. So a trade occurs at the price you already agreed to or it doesn't. Your maximum losses are controlled by you. I also like the conservative side of stock options (covered call and covered put options). You sacrifice control of your money or a security for several months for a payment now. This is much harder to game and computers have no advantage on time frames of months or longer.

End result

[Anonymous Coward]

We are well on the way to the ultimate end result of capitalism - a single computer that owns all the money in the world....

Average stock purchase held under a minute

[cpm99352]

Majority of trading (at least in the US) is computer. According to this, average length of time a stock is held is under 35 seconds.

The mainstream financial reporting in the US is a complete joke -- everyone fixates on the Dow, as though it held any meaning. At the end of each day, some "meaningful" reason is given for a less than 1% move, however automated trading never seems to be included.

Netflix joins the Dow??? Is that what this country is reduced to? No manufacturing, just service?

Meanwhile, SEC regulation is a total joke, insider selling is rampant, accounting is a joke...

But, if you're a retiree, where else can you hunt down returns? CDs are long dead.

Re:Average stock purchase held under a minute

[Desler]

No manufacturing, just service?

It's amazing that the US can be both the #1 manufacturer in the world and at the same time according to you have "no manufacturing" at all.

Re:Average stock purchase held under a minute

[Anonymous Coward]

But, if you're a retiree, where else can you hunt down returns? CDs are long dead.

They're getting into things that aren't appropriate (i.e. high risk or huge fees): stock market, gold, variable or indexed annuities (they're "guaranteed" returns don't include all the fees they nickle and dime you with.), junk bonds, and god knows what.

There was a New Yorker article a couple of years ago where the writer was invited to a party with all those hedge fund guys and big shots. With pride has asked what can he do with his $200,000 nest egg. They gave him the same old stale advice about diversification, risk, blah blah blah. When he pinned them down, they admitted for the little guy (less than 10 million liquid net worth - houses don't count), there's not much you can do. You're at the mercy of the markets - you're only chance is to go along with the current.

I've seen some guys do pretty well (win more than lose) with small caps and whatnot, but that's really their second job considering the time they spend on searching for opportunities.

Re:Average stock purchase held under a minute

[RightSaidFred99]

There is no manufacturing growth in the US, but we're still (for a little while) top manufacturer. But the post you replied to is correct in that sense - new manufacturing jobs aren't being created.

Re:Average stock purchase held under a minute

[Desler]

There is no manufacturing growth in the US,

Not true either. It is trivially easy to find articles with data to the contrary.

Re:Average stock purchase held under a minute

[0123456]

Not true either. It is trivially easy to find articles with data to the contrary.

Are those the articles where flipping a burger into a bun and sticking a piece of lettuce on top is counted as 'manufacturing'?

Re:Average stock purchase held under a minute

[Billly Gates]

Mod up.

Bush changed the rules and McDonalds workers are considered manufacturers.

We are the #1 manufacturer corporate headquarters in the world. The products are actually made in China by these American companies.

Re:Average stock purchase held under a minute

[umghhh]

Is US still #1 in manufacturing??? Well maybe it is but even if it were for how long would that be - other big countries grow fast very fast. Not that this is a good or bad thing it is just a fact - manufacturing output of US in long term has been in decline. If it were to fall further then it still has a long way down but that is how long term trend is looking like.

They put the servers in The Situation Room

[Dachannien]

In other words, the IT guys who maintain these servers all have greasy hair, don't wear shirts, and are total douchebags.

Re:They put the servers in The Situation Room

[TheL0ser]

May their hair be set aflame by the heat from the servers.

Re:They put the servers in The Situation Room

[couchslug]

"In other words, the IT guys who maintain these servers all have greasy hair, don't wear shirts, and are total douchebags."

Where do I apply?

Quite computers?

[maxrate]

In the 60 minute video, he goes on about saying something about how quite the machines are (exception being the air conditioning). Is there something I'm missing? I've been in plenty of server rooms and servers always seem to be noisy. Is he trying to contrast a traditional trading floor with the server room (being quite)? OR Are these some type of super new server that doesn't make any noise that I'm simply not aware of? The other thing I'm wondering about is the 65 micro (not milli) second times. What permits this incredibly fast trip time? Are they dealing with simply ethernet here or something else? Otherwise looks like a typical data centre - far more tidy that I normally encounter however.

Re:Quite computers?

[hitmark]

could be they are runking everything off ram and ssd so to minimize latency. Less moving parts equals less overall noise.

Re:Quite computers?

[Terrasque]

Quite what?

Sorry for being such an ass. I realize you meant "quiet", but seeing that typo over and over in your comment felt like someone was repeatedly stabbing my brain with a toothpick :(

Re:Quite computers?

[Anonymous Coward]

It's because these servers are basically next door to the exchange itself and have dedicated pipelines to it. When they say server farms in Jersey, that's only some of them. Many of the biggest clearing houses have their server farms right on Wall Street itself. In 1 microsecond light can travel 300 meters. In 65, it can travel 20km. So if the site is 6 miles away - say just over the bridge to Jersey - a round trip at the speed of light takes 65 microseconds.

Re:Quite computers?

[Jeremi]

In the 60 minute video, he goes on about saying something about how quite the machines are (exception being the air conditioning). Is there something I'm missing?

I'm not sure, but in the video the computers appear to be inside wire cages, in the center of a very large, warehouse-like room. I'd imagine that all the usually fan noise (etc) simply dissipates into the large space. In a lot of smaller "computer rooms", OTOH, the room is packed with computer equipment and thus the sound has nowhere to go and bounces off all of the walls. (By analogy, imagine a picnic in the middle of an otherwise-deserted soccer field, vs the same number of people packed into a small restaurant with glass/metal walls. Even if the people are making the same amount of noise in both cases, the restaurant will seem much louder)

The other thing I'm wondering about is the 65 micro (not milli) second times. What permits this incredibly fast trip time? Are they dealing with simply ethernet here or something else?

The fastest computer and networking equipment money can buy. I forget the exact name of it, but it's definitely not your standard gigabit ethernet from Frye's.

Re:Quite computers?

[Anonymous Coward]

fiber not copper cables

Re:Quite computers?

[LordNacho]

In the 60 minute video, he goes on about saying something about how quite the machines are (exception being the air conditioning). Is there something I'm missing? I've been in plenty of server rooms and servers always seem to be noisy. Is he trying to contrast a traditional trading floor with the server room (being quite)?

OR

Are these some type of super new server that doesn't make any noise that I'm simply not aware of?

The other thing I'm wondering about is the 65 micro (not milli) second times. What permits this incredibly fast trip time? Are they dealing with simply ethernet here or something else? Otherwise looks like a typical data centre - far more tidy that I normally encounter however.

I'm not able to watch the video, but I've actually taken a phone call from an engineer in one of the data centers mentioned in the NYT article. It sounds like he's standing in a hurricane. The aircon in the building and the fans on all the machines must be the source.

No idea why someone would call it "quiet".

Re:Quite computers?

[maxrate]

My goodness - how embarrassing - thank you for pointing this out to me. I'd like to blame it on a typographical error, but the truth is that the error is a hybrid between a 'muscle memory' effect and me being an non proof reading mental case. Thanks for pointing this out. Don't apologize - I couldn't agree with you more here. Sorry!

Re:Quite computers?

[maxrate]

You could be right - I just find the 1U and 2U stuff extra noisy - even if there are no hard drives and even if the room is 'cold' (keeping the RPM of the fans lower). Been in many data centers and I've never considered them to be quiet. Thanks for your insight.

Re:Quite computers?

[maxrate]

Yes, I thought this as well. I still think the CPU's and power supplies fans would be noisy (ac or dc). Good call on RAM and SSD.

Re:Quite computers?

[Anonymous Coward]

https://secure.wikimedia.org/wikipedia/en/wiki/InfiniBand

yes, but

[Anonymous Coward]

does it run linux?

Common View, Common Error

[istartedi]

If you can find a way to reduce bid-ask spreads without this kind of stuff, then I'll agree. Until then, I can't join the chorus of detractors.

With liquidity in the market, anyone who buys stock gets a narrow spread. Take liquidity out of the market, and you send us back to the dark days when stocks would trade at $1/8th spreads if you were lucky. $1/4 was common. Not only did you pay higher commissions, you paid the spread.

Unless you're pining for the days when you called your broker, paid him a percentage of the trade, and he placed your order in a market with a huge spread then you should be thanking the liquidity providers, not bashing them.

The current system doesn't hurt the little guy. The old system made it so the little guy wouldn't even think about it. I know, because I came of age when the old system was still in place for a few years. Buying in with a $1/4 spread on something trading for $10-$20, and then waiting for a significant percentage gain just to cover the spread??? No thank-you. HFTs? I LOVE them.

Re:Common View, Common Error

[DeadCatX2]

That's not how I read HFT. It sounds like skimming to me. Perhaps you could provide more details so I can understand what exactly you mean by "bid-ask spread", and how exactly HFT lowers it by holding onto stocks for a few hundred milliseconds.

Short selling, yeah, I can see how short selling can be useful for instigating corrections to over-valued stocks. But HFT? Seems a lot like "I can buy big, fast computers and a high-speed connection to the stock exchange, so I get to squeeze some value out of the stocks just before you buy them."

Re:Common View, Common Error

[e065c8515d206cb0e190]

No offense, but if you need a definition of bid-ask spread, you need to learn the basics before criticizing HFTs.

Re:Common View, Common Error

[iluvcapra]

To me a wide bid-ask spread is a small price to pay if the alternative is brokers using their access to the market information in order to extract rents from people who don't have such good information.

Re:Common View, Common Error

[Anonymous Coward]

That's not how I read HFT. It sounds like skimming to me. Perhaps you could provide more details so I can understand what exactly you mean by "bid-ask spread"

If you don't know what "bid ask spread" means, your read on HFT is worse than valueless- any proposal you make is far more likely to cause harm than good.

To put it another way, should we take computer advice from someone who asks "what do you mean by CPU?" (Yes- "bid ask spread" is that basic a concept).

Re:Common View, Common Error

[DeadCatX2]

No offense, but if you had given me a real answer, perhaps I might have learned something.

Re:Common View, Common Error

[diegocg]

The old system made it so the little guy wouldn't even think about it.

In contrast, with the new system you allow big boys to squeeze out your money just to get the illusion that the spreads don't exist.

Re:Common View, Common Error

[istartedi]

Perhaps you could provide more details so I can understand what exactly you mean by "bid-ask spread", and how exactly HFT lowers it by holding onto stocks for a few hundred milliseconds.

OK, let's say there's a hypothetical security X.

X BID 45 @$0.50 ASK 10 @$0.70

If you want to buy security X, you could try bumping up the bid to $0.55 and see if anybody will temporarily lower their ask that far.

The market for security X has terrible liquidity.

Now let's say somebody looks at this, and sees the awful liquidity. They say, hey, the market-makers suck. Let's do something better. They start "scalping" security X. They place bids at $0.55, then immediately flip for $0.65.

Now the market for security X has better liquidity. Somebody interested in X might be more inclined to buy, knowing that it doesn't have to rise too far before they can reasonably cover the spread.

The market-maker just bested the previous market maker. Market-maker A is shut out.... unless he can narrow the spread even further.

Now, the market-maker doesn't actually want to speculate in X if he can avoid it. It's in their interest to hold for as short a time as possible, and to make all their money off scalping or "skimming" as you call it. Without some kind of market-maker, the spreads are wider. You can hate the market-maker if you like; but try finding a better way to narrow spreads? I haven't heard of it.

That's how it works, in a nutshell. I'm glossing over a lot of detail, such as rebates and the different ETNs, and a lot of other stuff.

Note, I'm not belonging to an HFT religion. I'm just seeing it as "the worst system except for all the others". I don't believe in just turning these guys loose without regulation or oversight.

You asked for an explanation of how this stuff works, and I've given you my best understanding of it. I don't hold myself out as an expert. I'm just somebody who has been trading a bit, and watching markets since my teens...

Re:Common View, Common Error

[lysdexia]

http://en.wikipedia.org/wiki/Bid-offer_spread

Re:Common View, Common Error

[Estanislao Mart�nez]

If you can find a way to reduce bid-ask spreads without this kind of stuff, then I'll agree. Until then, I can't join the chorus of detractors.

It reduces the bid-ask spread by getting in between people who would have traded anyway, and giving each a worse deal. No benefit there.

Re:Common View, Common Error

[Anonymous Coward]

I object to your assertion that these systems add liquidity. They consume as much as (if not more than) they add. They are executing orders as much as they are adding to the book.

Volume != Liquidity
the same as
Volatility != Risk

Re:Common View, Common Error

[Anonymous Coward]

Perhaps you're just naive enough to be convinced you should be happy that now you can mostly avoid paying the fees for placing limit orders?

Or perhaps you're a would-be day trader who got tired of having to hold a stock long enough for it to rise more than 2.5% (your hypothetical $1/4 spread on a stock trading for $10)?

HFT never got rid of spreads, it only made them appear gone by advertising artificial** prices and guaranteeing that the true spread difference always goes into the pockets of the folks who run the HFT systems. Before HFT you could pick up profit from the spread at least part of the time.

** its "economic theater" of much the same spirit as the US Govt's revised inflation calculations made it appear that inflation is super-low by artificially removing the most-inflationary items from the CPI. Did you notice the recent announcements that China is having trouble with inflation? Did you read the details and see that the specific items that caused China's inflation numbers to pop higher than desired are the same specific items (housing, heat/transportation fuel) that the US Govt started excluding from its inflation numbers a few years back?

Re:Common View, Common Error

[istartedi]

That's a compelling argument.

Bid-ask spread is readily quantified, and readily available. Any inequity due to HFT might be harder to quantify.

In highly liquid, HFT dominated markets, it's possible for the price to be unfairly set. The big problem? How do we know what a fair price is?

There are a lot of standard numbers you use to evaluate stocks: P/E, etc. At the end of the day though, it always boils down to market price.

The idea that the old bid-ask spread is "hiding" in the new market is possible; but how do you quantify it? Where would it go? I'm purely speculating here (no pun intended) but perhaps it went to volatility. It would be interesting to go back and look at real data from HFT vs. non-HFT markets, and see if there is more volatility in them.

Note, volatility doesn't hurt small investors unless they get stopped out. Stop-loss is a double-edged sword, and options have problems of their own. To reiterate, I didn't say HFT was perfect; I just don't think it's the demon that some make it out to be. A buy-and-hold investor isn't affected by volatility in the short run at all--certainly not the kind that HFTs might cause. An HFT can only drive the price of International Buggie Whip to a PE of 1000 for so long before they get burned.

Or, more simply, are market-makers "taxing" trades more now or less? Certainly there is more trading VOLUME now; but is the percentage going to HFTs greater or less than the percentage going to brokers/specialists under the old system?

If HFTs are making $1 billion on a trillion trades, while brokers/specialists made $500 million on 200 billion trades, which is more fair?

Plainly, further study by guys whith degrees different than mine (and actual jobs doing the studies) is needed...

Re:Common View, Common Error

[DeadCatX2]

Thanks for the example; I hope you are rewarded by being modded up. You may not be an expert, but I'm not even an amateur, so at least I can digest your example.

Reducing the spread increases liquidity...check. What about stocks that pay dividends; you're supposed to hold on to them long-term, so does a narrow spread benefit those stocks at all?

Re:Common View, Common Error

[istartedi]

What about stocks that pay dividends; you're supposed to hold on to them long-term, so does a narrow spread benefit those stocks at all?

You could just own the stock on the ex-dividend date, sell the next day, and still get the dividend. I've never tried it. In theory the market should know about this and move the stock so as to make it a pointless strategy. I'm sure that doesn't stop people from trying.

Anyway, I'd tend to say that for people buying and holding dividend-paying stocks for the long term, the impacts of HFT don't matter one way or the other.

Re:Common View, Common Error

[Compuser]

Why do we need low spreads? Seems like if spreads are high then you will think twice about investing which is great as it discourages speculation. And if indeed, maximizing spreads is in fact desirable to ensure maximum consideration prior to investment and hence maximum system stability, then why do we need market makers in the first place?

A second question. When all market makers run HFT algos, does this not raise the possibility where machines constantly flip securities to each other without them being passed on to an actual investor? Because that would be bad. It would inhibit meaningful price discovery for one thing and consequently, resource allocation in the greater economy would be suboptimal.

Re:Common View, Common Error

[Estanislao Mart�nez]

I object to your assertion that these systems add liquidity. They consume as much as (if not more than) they add. They are executing orders as much as they are adding to the book.

And what's worse is that every time I look into these schemes, they seem to be predicated on detecting matching buy and sell offers quickly enough to get in between them before the market pairs them up. That is, the HFT only kicks in when a trade was going to happen anyway. That not only doesn't add liquidity, it assumes that the liquidity is already there.

Re:Common View, Common Error

[LordNacho]

I've worked in markets for years, though nobody can really can themselves an expert in something so complex.

As for dividends, there's no requirement that you hold on to high-div stocks for a long time. Also, and this is critical, you know exactly when that dividend is coming out of the price. So, basically, it completely doesn't matter whether the stock is paying a high dividend or not, since the liquidity providers know this. Having a narrow spread is good for you no matter what you're buying.

BTW, you might like to consider that ALL prices can be considered as bid/ask spreads. It's just that in your everyday life, people tend to only show you ONE side. In a supermarket, all the prices are Walmart's ASK price. Walmart got your potatoes by BIDDING for them from the wholesaler. They pocket the difference (bid-ask spread) just like a market maker.

When you buy a house, you are BIDDING (though confusingly this is colloquially called an offer on a house). If the owner is wanting 100K, and you are willing to pay 95K, one of these things will happen: 1) You cross the spread to LIFT THE OFFER 2)He crosses the spread to HIT THE BID 3) One of you moves their price and either 1 or 2 happens. The guy who crossed the spread is the aggressor, and he basically has swapped 5K for immediacy, ie to make the deal happen now rather than wait.

Re:Common View, Common Error

[TurtleBay]

Higher spreads increase the amount of money that the market makers (Wall St) make off the trade and decrease the amount of money that goes to the investors (the buyer and the seller of the stock).

Re:Common View, Common Error

[LordNacho]

Analogy for why you need a market maker:

There's a market where farmers come to sell pigs, and butchers come to buy them. Unfortunately, the farmers and the butchers are not there at the same time. They might arrange a market day, which concentrates the trading in certain periods. Fine. But that leaves risk. What if market day is in a few months, and farmer doesn't know whether it's worth making some more pigs? Maybe he'll keep safe and decide not to. Result: fewer pigs, even if butcher actually wanted to buy them.

So, let's keep the market open. But farmers and butchers won't know when to show up? What if I go to market, and no butchers are there? Same problem as before... take less risk.

Enter the market maker. He doesn't know how to farm or butcher, but he does hang out at the market all day. He has a good idea of how many pigs are needed, because each time a real farmer/butcher comes in, someone is asking around about the pig price. So he uses his own money to buy pigs from farmers and sells them to butchers, based on what he thinks the balance is. So now, farmer guy can offload his pigs, and butcher guy can buy them, without having to coordinate with each other. They just talk to the middle man.

Now, the middle man has to manage the pig inventory. What prices does he make? Well, again, he thinks about risk. If he makes a wide spread, he'll make more money on each turn. But that will discourage the farmers/butchers, because they lose more from when they do their business. But if he tightens, he'll make less. He'd have to have enough business to compensate.

Enter middle man 2. He spots the same opportunity. Interestingly, now the two middle men need to trade with each other. And also, their trading doesn't need to be governed by the external customers.

Re:Common View, Common Error

[LordNacho]

A wide bid-ask spread IS a form of rent. Just look at the FX shops in any airport.

Re:Common View, Common Error

[Anonymous Coward]

http://www.investopedia.com/articles/02/110802.asp

Re:Common View, Common Error

[iluvcapra]

It's only a rent in this case if the the spread results from an information asymmetry. A bid-ask spread happens in the open as a part of the quotation, so it's (ideally) subject to competition and market makers take the spread as a legitimate cost of doing business. Better to make people pay for liquidity in the open than have machines crank it out in the dark, where they can suddenly and mysteriously withdraw it.

Re:Common View, Common Error

[LordNacho]

Well, it's the FX shops raping you because you're in an airport terminal, with no access to sensible prices. But I get your point.

With regard to "in the open" prices, the first thing human market makers do in a crisis is shut down. Just like machines, except humans are slower.

Re:Common View, Common Error

[LordNacho]

What do you mean by this? The market enforces rules as to which orders are first in the queue.

Re:Common View, Common Error

[Anonymous Coward]

Or SETS.

Re:Common View, Common Error

[Anonymous Coward]

Bid-ask spread, as its name suggests is the difference between the bid and ask prices for a stock, i.e. what people are willing to pay and how much they demand to sell. To demonstrate how this works, lets look at a simplified model - suppose there are 10 people looking to sell one share and 10 others looking to buy one share, each with different prices that they are willing to pay (bid) or sell for (ask). If there is a bid price higher than an asking price, a trade occurs. If you just bought the share, commissions and taxes would mandate that you must ask for more to immediately resell it. Similarly, it makes little sense to re-buy what was just sold for more than you received.

This reasoning means that if ask and bid prices for each individual remained static, we would eventually arrive at a point where all ask prices were higher than bid prices. At this point, the market bid price is the highest of the bids and the market ask price is the lowest of the asks. Now we relax the model a little and allow other investors to enter one at a time. Some may be willing to pay more than the current market bid or demand less than the current market ask, in which case a trade would occur. What an HFT would do in this situation is step in and offer to sell for marginally less than the ask price and buy for marginally more than the bid price. This lowers the market ask and raises the market bid, decreasing the bid-ask spread. When a new investor enters, he is more likely to find a price that is acceptable to him and will always trade with the HFT and receives a slightly better value than would otherwise exist. The HFT makes money due to bid-ask spread between its purchases and sales. Now we speed up the model. A patient investor (one using a limit order for instance) doesn't need the market making function of HFTs, but those who for whatever reason want an immediate trade benefit by having better offers (particularly as HFTs spread). The tradeoff is a small marginal loss of funds for being able to immediately sell the share. Compare this to real estate where foreclosure sales demonstrate just how much of a premium there is for rapid sale of real goods, and you come to appreciate how little HFTs cost for their benefits.

Re:Common View, Common Error

[swb]

While what you say makes sense, I think the entire philosophy is wrong. I think it's fueled by short term profiteering vs. investing for a profit.

Wider spreads and a more limited market were a disincentive for small-value, short-hold stock trading; they really didn't have any impact on long-term investing in stocks, which was more about buying, holding and collecting dividend checks. Mutual funds had to be more straightforward "baskets of stocks" and not complex, opaque investment vehicles with more churn than a pound of butter.

Basically, the market was more like what it should be for -- a place to invest on longer time horizons and a place for business to raise capital necessary to do something, not a quantitative casino where the house takes bets, makes bets and picks the winners.

Maybe I'm just nostalgic, but I can't help couple American economic decline with all of the short-term thinking associated with HFT, derivatives and all the other smoke and mirrors investing Wall Street has to offer.

Re:Common View, Common Error

[Anonymous Coward]

WTH is the "spread" isn't that a term used by people who gamble? Tell the bookie I can cover the spread, is a random line from a movie I believe.
But the way you used it I would guess you mean trading fees.

Re:Common View, Common Error

[smellotron]

I must agree with the GP that if you don't understand something as fundamental as the bid-ask spread, your personal opinion on HFTs (or any other "classes" of traders) just isn't very useful or valid. To put it as a car analogy, it's akin to arguing with your car mechanic, "I don't know what an interference engine is, but I just don't see the value in a timing belt."

You can read the book Trading and Exchanges if you really want to get into gritty details. Otherwise, if there's ever a term you don't understand I suggest searching for it on investopedia.com. Both of those are very good resources for learning more about the markets.

Re:Common View, Common Error

[istartedi]

America, particularly the western states, were much more inclined to gamble back in the 'ol days. Gold mining claims, striking out into the territories to reap a harvest or get reaped.

When guys weren't gambling their lives in mines or on horseback, they were gambling around a card table (OK, maybe not as much as the movies depicted, but the West was full of risk).

Shit. The whole country was a gamble. If anything, I'd say the introduction of high-stakes financial poker into our living rooms represents a return to Amercian values; but that's just one man's opinion.

Re:Common View, Common Error

[smellotron]

BTW, you might like to consider that ALL prices can be considered as bid/ask spreads.

I've already posted so I can't moderate, but these are great examples.

Re:Common View, Common Error

[istartedi]

Yay for pork bellies! :)

You just reminded me of one of my favorite stories--the story of why onion futures were made illegal, and AFAIK they still are. Markets are great; but they don't always work.

Re:Common View, Common Error

[sjames]

The thing you missed though is why was it worth it to the market? If I'm bidding $0.50 and nobody's willing to meet me there, I can either give up on it or try $0.55. Given your example, someone willingly meets me there and all's well. Unless, of course someone swoops in before I can even finish keying it in and scoops it up at $0.55 and tries to flip it at $0.65 right away. They just effectively jacked the price up (since there's no way they're going to sell it at cost) and contributed nothing to the process.

Frankly the whole thing sounds like that old hard line Communist wet dream of the master computer running the economy except that there's not even a pretense of doing it for the public good.

Re:Common View, Common Error

[dookiesan]

One person is willing to sell for $10.25 and this is the lowest price out there. Someone else is willing to pay $10 and this is the highest offer.

I have no idea why such a spread would be sustained if HFTs were deprived of millisecond execution times.

Re:Common View, Common Error

[Anonymous Coward]

Fix your signature:

http://wiki.answers.com/Q/Is_the_saying_'all_intents_and_purposes'_or_'all_intense_purposes'

Re:Common View, Common Error

[the way]

If you can find a way to reduce bid-ask spreads without this kind of stuff, then I'll agree. Until then, I can't join the chorus of detractors. ...
Unless you're pining for the days when you called your broker, paid him a percentage of the trade, and he placed your order in a market with a huge spread then you should be thanking the liquidity providers, not bashing them.

You just answered your own question - the reason for the high bid-ask spreads in the old days was because of how the industry was structured: old-boys networks that were the only ones that could place trades, high commissions, plus higher minimum spreads as a percent of price (on average).

Now that we have online trading, commissions can come down, so spreads can decrease. However, they don't have to decrease to nearly zero - automated trading systems that take the difference for themselves as profit aren't benefiting anyone. Placing trades in random order at one second intervals would get rid of the HFT madness, whilst still maintaining liquidity (and then the liquidity reflects people who really want to trade in the stock - not algorithms gaming the system).

Re:Common View, Common Error

[LordNacho]

They don't always work, true. But tinkering with them isn't too easy either.

Re:Common View, Common Error

[TheLink]

What do you mean by this? The market enforces rules as to which orders are first in the queue.

Unfair rules in this context. They get to be 30 milliseconds ahead of everyone:
http://www.nytimes.com/2009/07/24/business/24trading.html
http://www.nytimes.com/imagepages/2009/07/24/business/0724-webBIZ-trading.ready.html

They can also post orders AND cancel them before others can go through with the transaction. So if you have a simple automated system - they can figure out what your minimum/maximums/rules are.

Quote first link: "High-frequency traders often confound other investors by issuing and then canceling orders almost simultaneously. Loopholes in market rules give high-speed investors an early glance at how others are trading. And their computers can essentially bully slower investors into giving up profits -- and then disappear before anyone even knows they were there. "

Re:Common View, Common Error

[Politburo]

The government didn't exclude anything. If the media wanted to report CPI including food and energy, nothing is stopping them, you, or anyone else.

Re:Common View, Common Error

[istartedi]

Yep. In the case of onion futures, they probably could have just made some rules to prevent corners instead of killing the whole market. AFAIK, modern futures markets have rules designed to prevent corners. The Wiki article also said the son of one of the farmers hurt by the onion fiasco would actually like to see the futures market come back. Apparently, he ran into a situation where being able to hedge his crop (as other producers are able) would have been handy.

And mankind became slave of his own system

[Anonymous Coward]

First mankind created computers, to work for him the dull work of counting money.
Now these systems who were behind one of the major financial crisis; (its a a bigger secret then wikileaks).
Now these systems will define the value of money, the value work; keeping in respect the unbalance of this world.
The poor stay poor and their labor is cheap; and the rich steal their resourcers by the power of money.

Micro trading, is the most dangerous form, of the spirit that keeps this world devided; making the rich richer, and the poor loose everything..
To act with no moral in miliseconds; resulting in job loss; resulting in families to be destroyed; resulting in countries going bankrupt.
Machines dont understand love, dont understand what the real world is; and would do whatever it takes to make more money; trade landmines, or play pyramid games with pensions..

They do understand mankinds faith.. to work as a slave for profit, without knowledge of love, art, music, it will put mankind to 100% labor and work till its dead.

i've nothing to make this sound like a joke, we created it and allow this happen.

The most strange thing is however that if resources where devided equal, and we would share knowledge there is more proffit in life to gain; how fast would we cure current diseases; if we where not distracted by how many money goes to HIV or Malaria, Yellow fever and others; but where just united. There would even be food enough...

 

Fiat Lucri...

[modecx]

Presto Profito! These must be talented wizards, with the a wave of a hand and the utterance of some arcane terminology, the heavens crap monies right into their bank accounts! Let's see Harry Potter do that.

Re:Fiat Lucri...

[maxume]

Well, Harry Potter did crap a lot of money right into J.K. Rowling's bank accounts.

Re:Fiat Lucri...

[modecx]

Hrm. I suppose you're right! I guess ol Harry has a future in stocks after all.

My Proposed Solution for HFT BS

[Anonymous Coward]

Rule #1: All securities MUST be held for at least 24 hours.

Rule #2: Capital gains are now taxed according to the following scale:
Ultra Short Term (Less than 1 Week): 90% Tax
Short Term (1 Week to 6 Months): 50% Tax
Medium Term: (6 Months to One Year): 35% Tax (The Maximum Current Short Term Rate)
Long Term: One to Five Years: 15% Tax (The current long term rate)
Very Long Term: Over Five Years: 5% Tax

Rule #3: You can only claim capital losses on securities held more than six months.

The Stock Market should not be like Vegas. Rules need to be in place to provide incentives for long term investment.

pfft

[Charliemopps]

Change the rules. You can't sell a stock within 24hrs of buying it. There is absolutely no benefit in the hyper trading of stocks like this. Computers are just going to get exponentially faster until trading is so fast the entire market could crash before regulators could do anything about it. Hasn't it happened several times already and they had to roll back trades for several hours because the market was so screwed up?

Re:pfft

[TheWizardTim]

No, we should not limit what people can do with a stock. We could however make it expensive to trade this way. Change capital gains to (Less then 1 day, 50% capital gains.) (1 day to 1 month, 40% capital gains.) (1 month to 2 years, current short term capital gains.) (2 year - 20 years, current long term capital gains.) (20+ years, no capital gains.) This would allow people to trade less then a day, and others trade real long term. I like the idea of making it expensive to do rapid short term trades, but let people who invest real long term, 20+ years, get a strong benefit for it. The stock market should not be run like a casino.

Re:pfft

[jonbryce]

I think you will find stockbrokers want to sell their shares as quickly as possible. They make their money from trading, not from holding and if they had to carry the risks of ownership, then buying and selling shares would become a lot more expensive.

Re:pfft

[Anonymous Coward]

No - Put a federal sales tax on all stock purchases. End of problem.

Re:pfft

[TheSync]

Change capital gains to (Less then 1 day, 50% capital gains.)

Or perhaps corporations should consider offering a class of less-volatile shares. As far as I know, there is nothing stopping a company from issuing a class of shares that, for example, can not be sold for 24 hours after purchase. Many executives and founders are issued shares with significant restrictions on sales.

I would suspect such a class of shares would in general trade for less than the more flexible shares.

Do $0.01 trades matter to non day traders?

[perpenso]

... Take liquidity out of the market, and you send us back to the dark days when stocks would trade at $1/8th spreads if you were lucky ... Unless you're pining for the days when you called your broker, paid him a percentage of the trade, and he placed your order in a market with a huge spread then you should be thanking the liquidity providers, not bashing them ... I know, because I came of age when the old system was still in place for a few years. Buying in with a $1/4 spread on something trading for $10-$20, and then waiting for a significant percentage gain just to cover the spread??? ...

I believe you are misrepresenting the old system to a degree. You did not have to call a broker, the do-it-yourself fixed-cost-trade online brokers were available long before decimalization. I also don't recall much difficulty getting something at the 1/8 limit I wanted. Now I wasn't doing some kind of twitchy trading, if I put in a limit order and it took an hour or two to close that was no problem. I don't think 1/8 vs 0.01 pricing makes much different unless you are frantically day trading.

So in a few years...

[Fuzzums]

... it will become important to model the behaviour of this software and not how "the financial market" behaves.

stocks != livelihood

[Anonymous Coward]

"part of your wealth, your livelihood"

I'm sorry, but I was taught: never gamble that which you cannot afford to lose.

Wealth in the stock market is NOT my livelihood

Re:stocks != livelihood

[iammani]

If you are invested in pension or mutual funds, you do have wealth in the stock market. If you are not, you dont speak for the majority, and your opinion doesn't matter.

TradeNet

[Anonymous Coward]

From Bankinator 2: Expiration Day:

The Bankinator: The Tradenet Funding Proposal is passed. The liquidity center goes on-line August 4th, 1997. Human decisions are removed from stock trading. Tradenet begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, August 29th 2012. In a panic, they try to pull the plug.

Sarah Connor: Tradenet fights back.

The Bankinator: Yes. It invests in then crashes certain stocks. Vast fortunes are lost in hours. People jump off skyscrapers. Pensions become worthless. Oil and gold become havens for a while and commodity prices become sky-high. All major economies die overnight.

You wouldn't get it...

[Anonymous Coward]

It's a Jersey thing!

how shorting a stock works

[hildi]

1. you borrow stock 2. you sell it 3. you wait till the price drops 4. you buy some of the stock (even if it is still falling, you never know when the bottom will hit) 5. you return the 'borrowed' stock to the original lender between step 2 and step 4, your income was high_price*number_of_shares, minus low_price*number_of_shares

Re:65 microseconds

[TurtleBay]

a large part of what these companies are paying for is that they are only two hops from the servers for the stock exchange. The idea is that the router of that data center is directly wired to the exchange (which is in the same or an adjacent building) without any ISP in between.

Tobin Tax Now!

[transami]

Tax all trades ~1%, problem solved.

Re-couple Market Access With Market Making

[tyen]

The standard explanation proffered by the HFT owners and customers is they "add more liquidity". This is repeated so many times that laypeople buy it; see typical comments in this thread like "we have traded wider spreads for higher instability". This is not the entire story: the liquidity is for them, not for you . That there is sometimes a spillover liquidity and spread improvement for participants in the wider market is merely a convenient observation suitable for PR. The past and ongoing flash crashes demonstrate that when the liquidity trades against them, they pull this vaunted liquidity quicker than you can blink, literally. They're not going to leave money on the table supplying liquidity into the market if they don't have to.

Another oft-made claim is "anyone is welcome to do what we do, there are no barriers to entry". That is not quite the entire story as well. The defining feature of an HFT firm over the retail investor apart from scale (you need accredited investor-scale financial depth just to ante up the money to the exchange to cover their risk for you fracking up your code and making market on your fracked up orders they then have to make good upon) is access, as the articles this story links to amply documents. They are quite different from most market participants. While it is true that one doesn't have to have special institutional privileges and access to buy these newfangled digital-age "exchange seats", and "merely satisfying" some financial and technical criteria make these seats putatively easier to obtain than the old seats, make no mistake about it, they are more privileged than the old school NYSE exchange seat holders: they enjoy special access to the markets that "non-seat holders" do not, namely preferential positioning in the order flow inspection pipeline, or put another way, they enjoy market making access without market making responsibilities. Just because you no longer have to have a hallowed name descending from the Mayflower, a family history intertwined with the exchange, and an imposing granite edifice for offices to qualify for an exchange seat that buys access to the order flow doesn't mean that preferential access is open to everyone. The day the exchanges open up the HFT level and quality of tick access for the same price as 15-minute delayed ticker quotes, would be the day that I withdraw this observation.

If you chafe at these new special breed of privileged market participants, then an old school remedy is still available: with privileged market access, comes market making responsibilities and market making regulatory oversight. Perhaps not as much responsibility as the exchanges, but definitely more than those without the preferential access, commensurate with their impact upon the market as shown by the flash crashes. Let them have the special access, but make good on the liquidity and spread claims with regulatory enforcement; that is, they continue eating at the trough even when the liquidity and spread moves against them. It didn't stop the old school market makers from coming up with different licenses to print money, so they'll still make great bank (though they'll bitch like a platoon of coked-up noob IB's at Penthouse for having to run through regulatory hoops that didn't exist before, instead of spending that time cranking the next batch of algorithms onto FPGAs), but coupling privileged market access with market making responsibilities did truly impart long-term benefits to participants in the wider market. Arguable if the benefit was proportional, but as long as we will tolerate differential access, we might as well at least maintain the marginal benefits of status quo ante, eh?

Re:Re-couple Market Access With Market Making

[alexmin]

When retail investor enters orders into his Schwab/Ameritrade/Interactive Brokers web portal, guess where those orders go? Yep, his broker colo facilities in Mahwah (NYSE), Carteret (NASDAQ) or Weehawken (ARCA/BATS). Main difference is that those orders get exercised by broker-owned systems and not customer's. Want your's gear in place? Power, cooling are not free as you probably know so be ready to pay up beyond $10/month account maintenance fee.

Anyway, you are missing the point. Investment in stock market does NOT require frequent executions. It is about buying when the stuff is cheap and then holding long time (like many years long) and then selling when you need money (not when it's expensive, that's speculation.) Speed is not important, valueing correctly to know when stuff is cheap is paramount.

I am involved in trading for living but did not touch my personal account in many-many months.

Re:Re-couple Market Access With Market Making

[Red Flayer]

It is about buying when the stuff is cheap and then holding long time (like many years long) and then selling when you need money (not when it's expensive, that's speculation.)

I'm sorry, that's so far off the mark it's not funny.

Joint stock corporations were founded entirely on the concept of speculation and risk. Stock markets do not exist a place to dump your money and hopefully have it grow until you need the cash.

Stock markets exist to provide liquid capital to ventures; what they offer the people providing that capital is the chance to make speculative profits and a diversification of risk. This is the fundamental purpose of stock exchanges.

Speed is indeed important, because it allows the investors (speculators) to properly evaluate and buy/sell stocks.

The problem is not that speed is important -- the problem is that a small group of people/firms have an advantage over all other people participating in the market, because not everyone has equal ability to trade at high speeds. This allows them to engage in rent-seeking behavior that is detrimental to the market at large.

It used to be that having a scoop (that is, useful information before anyone else) gave the biggest advantage to traders on the market. Then we banned insider trading, so people would all be acting on the same public information. Now, that scoop is less relevant for another reason -- you may make some profits off of information asymmetry, but the HFTs are going to take most of your profits simply because they can trade faster than you. They can undercut your put or overbid your ask, and make money on both sides because they see the transactions before they happen.

I think I got a bit off-track there; what you refer to is not investing in the stock market. It is investing in single companies, sectors, or funds. And even if this is what you do (like the original investors in joint-stock ventures), you should still seek to sell when the value of your stock is the highest weighed against the risk you find acceptable.

This brings me to my main beef with HFTs. The entire purpose of stock investment, at its core, is to maximize your investment returns while minimizing the risk to your capital. If you owned a ship in the 1500s, you could be wiped out, financially, by a storm. If you owned a 1/16th share of 16 ships, a storm wiping out a single ship would reduce your returns, but you'd still have assets equivalent to 15/16 of a ship. So in exchange for reduced risks* on each single venture (a ship, in this example), you'd see slightly lower profits from successful ventures. Meanwhile, the people who initially sold those shares to the owners of the joint-stock ventures would have some amount of liquid capital, which they would use to build or acquire the ships, crews, etc, they promised. This was the primary purpose of the original joint-stock corporations -- provide capital to businesses by allowing people to invest at less than the full cost of the venture, and spread risk among investors.

Of course, HFTs take profits with very little risk... this is why they are bad, IMO, for the stock market. Stock markets are about trading risk for profits. HFTs reduce profits for risk-takers, thus making, in the long run, less capital available to risky ventures.

It's funny, though -- you never hear a HFT firm talk about assumption of risk, they only talk about providing liquidity. Liquidity benefits stock speculators, but it doesn't really benefit the actual businesses who produce goods and services and provide meaningful employment.

* discounting systemic risks, which apply to all individual ventures in a class -- and as such, are incidental to the concept of risk diversification within a class.

Re:Re-couple Market Access With Market Making

[tyen]

They are called mini flash crashes, but only because of new circuit breakers put in place after the big one. I am a more old school, Graham-style market participant so I don't use GTC's. Most of my exits are measured in multiple years from my entries, and I'm switching over more to bonds instead of participating in secondary markets as a response to the principal agent problem. Even so, the vig HFTs and other participants take out of my trades is buried in the rounding error on those duration scales, unless it is a losing trade of course. As long as my participation helps me stay even with inflation, I'm content to focus upon my business and make money the old fashioned way.

Re:Re-couple Market Access With Market Making

[alexmin]

The problem is not that speed is important -- the problem is that a small group of people/firms have an advantage over all other people participating in the market, because not everyone has equal ability to trade at high speeds. This allows them to engage in rent-seeking behavior that is detrimental to the market at large.

To paraphrase: the problem is that a small group of people who have these new fangled things called automobiles have an advantage over al other people on the streets because not everyone has equal ability to move at high speeds This allows them to engage in behavior that is detrimental to horse-driven carriages and pedestrians at large.

Question: are we better off in the past 100 years due to participation in the risk-taking behavior called "driving"?

freedom of choice

[Anonymous Coward]

one does not have to put one's money at risk.

one has freedom of choice.

one loses one's money because of oneself.

one is where one is because of one's choices.

if one makes money in spite of high frequency trading then what is the cause for others to lose money.

losers lose money trading because they lack a complete trading plan.

a trader trading a complete trading plan continuously makes more money than one loses

stay out of the market if you do not have a complete trading plan.

from a bedroom trading room with a retail account my trading buddy went from $50,000 to $6,000,000 to $600,000 to $10,000,000 now at $5,000,000.

twittering as stocktradr

Re:65 microseconds

[maxrate]

Private line stuff I guess. Probably GigE fiber or better. I went on Savvis.com website a few years ago and they mentioned ultra fast stuff for Wall Street. I just dismissed it as low latency Internet - sounds like I may be wrong.

Secret Location

[Anonymous Coward]

Could this be the secret location of their data center that the NYSE didn't want to have divulged by 60 minutes? You know the one I found in 5 minutes of searching?

http://goo.gl/maps/bF5d

adjusted to remove the screams

[Anonymous Coward]

..take out the budget busting wasteful and hideously ugly war machine manufacturing, the blood profits crap,

Now..how does manufacturing look in the US again?

WTF planet are you on? Been to any stores lately?? Do you really BELIEVE the government BS statistics on shit? Are you really that dumb? You are a voting age adult, correct?

I bet you believe the government's unemployment figures too, and the cost of living numbers, and all their other drivel.

The US is manufacturing high speed hideously bloody screaming painful irrational and demonic DEATH machines..and not much else, not anything like we used to, and I sure as shit am old enough to remember when most everything we had, including all electronics, was domestically manufactured. Fucking everything. Every.Thing.

I bet you aren't old enough to remember that, else you wouldn't try to pass that propaganda off as gospel.

We've turned into a nation of drooling drunk football violence addicts, pasty faced pedantic violent video game and ritalin addicted girly men nerds, government alleged "workers" and a war and finance skimming machine and that's about it. We don't make jack shit anymore, just assemble a few things that are made elsewhere. A few exceptions, but 99% of our industry has vamoosed except for war and big banking "products". Yes, they call that horseshit "products", "financial products". Maybe that's what you mean by "manufacturing". State lottery scratch off cards "products". Or like the other poster said, and it is true facts, they call assembling a burger "manufacturing" now.

Please stop spreading government/wall street economic lies/propaganda, stuff that is easily debunked by anyone rational using their eyeballs and going to ANY FUCKING STORE in the US. We don't give a shit how many articles you can "point to" by those industry shills, egads man, go outside into the real world and TAKE A LOOK!

The only problem with the whole mess

[Anonymous Coward]

The only problem with this whole mess, is that the algorithms are based on probability. You calculate the odds of something happening, and then calculate the eigenvector of the reverse happening, and then estimate, based on standard probability theory that the thing is going to happen. So if things are going south, you calculate the reverse of that, then look for when it looks like its going to be that thing, then bet according to the probability. The only problem with it is that normal probability usually happens, but there are a lot of things that can screw up the natural order, and then standard deviation and normal probability fly out the window. Example: leaves on a tree will fall in a uniform pattern in autumn so that a higher number of leaves will fall closer to the trunk, tapering down to fewer leaves at the outer branches. ....except that squirrels and cats and the wind and rain and dogs and cats scatter the leaves so that the uniform distribution fails, and how many other ways are there to wreck the standard distribution of leaves besides dogs and cats and wind and rain...? And so it goes with algorithms that calculate the probabilities of stock. Except that the problem is much more complicated than just leaves, there are more things affecting it than dogs and cats and the wind; millions more, and when things are calm and 'the market is stable' then everything is OK, but when the market is 'unstable', you get better odds using the random function on you pocket calculator than the data centers full of hardware.

Jersey?

[NCTRNAL]

Please tell me Snooki and The Situation were not involved.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Comment removed

[account_deleted]

Comment removed based on user account deletion